Prevention of menstrual attacks of migraine: a double-blind placebo-controlled crossover study.

OBJECTIVE: To assess the effect of perimenstrual estradiol supplements on menstrual attacks of migraine associated with estrogen withdrawal. METHODS: Women with regular menstrual cycles and menstrual migraine or menstrually related migraine completed an initial three-cycle assessment confirming eligibility for a six-cycle crossover study using estradiol or placebo to prevent menstrual attacks of migraine. Women collected early morning samples of urine daily for laboratory assay and used a fertility monitor to identify peak fertility associated with ovulation. Estradiol gel or placebo was first applied on the tenth day following the first day of peak fertility and continued daily until, and including, the second full day of menstruation. Women kept a daily migraine diary and continued their usual treatment for migraine. The main outcome was the number of days during gel use on which a migraine occurred. RESULTS: Data from 35 women were available for a paired analysis. Percutaneous estradiol was associated with a 22% reduction in migraine days (RR 0.78, 95% CI 0.62 to 0.99, p = 0.04); these migraines were less severe and less likely to be associated with nausea. This was, however, followed by a 40% increase in migraine in the 5 days following estradiol vs placebo (RR 1.40, 95% CI 1.03 to 1.92, p = 0.03). CONCLUSION: Although perimenstrual percutaneous estradiol showed benefit during treatment, this was offset by deferred estrogen withdrawal, triggering post-dosing migraine immediately after the gel was stopped. Further work could assess if this could be avoided by extending the duration of treatment with estradiol.

Incidence of migraine relative to menstrual cycle phases of rising and falling estrogen.

OBJECTIVE: To investigate the association between urinary hormone levels and migraine, with particular reference to rising and falling levels of estrogen across the menstrual cycle in women with menstrual and menstrually related migraine. METHODS: Women with regular menstrual cycles, who were not using hormonal contraception or treatments and who experienced between one and four migraine attacks per month, one of which regularly occurred on or between days 1 +/- 2 of menstruation, were studied for three cycles. Women used a fertility monitor to identify ovulation, conducting a test each day as requested by the monitor, using a sample of early morning urine. Urine samples were collected daily for assay of estrone-3-glucuronide, pregnanediol 3-glucuronide, follicle-stimulating hormone, and luteinizing hormone. All women kept a daily migraine diary and continued their usual treatment for migraine. RESULTS: Of 40 women recruited, data from 38 women were available for analysis. Compared with the expected number of attacks, there was a significantly higher number of migraine attacks during the late luteal/early follicular phase of falling estrogen and lower number of attacks during rising phases of estrogen. CONCLUSION: These findings confirm a relationship between migraine and changing levels of estrogen, supporting the hypothesis of perimenstrual but not postovulatory estrogen "withdrawal" migraine. In addition, rising levels of estrogen appear to offer some protection against migraine.